July 3, 2025 • 29 mins
Interested in being a guest? Email us at admin@evankirstel.com
The technology in your pocket—sensors in your everyday smartphone—is quietly revolutionizing road safety in ways most of us never realize. In this eye-opening conversation, MIT professor Hari Balakrishnan shares the remarkable journey of Cambridge Mobile Telematics (CMT) from university research project to global safety leader.
What began as a simple idea to use mobile device sensors to measure driving quality has evolved into technology that's prevented over 90,000 crashes and saved 50,000 lives worldwide. Balakrishnan reveals how CMT's DriveWell platform fuses data from phones, windshield-mounted tags, connected vehicles, and dash cams to create a comprehensive picture of driving behavior that's transforming the insurance industry.
The implications are profound. For the first time in insurance history, companies can price policies based on how people actually drive rather than demographic assumptions. Young drivers no longer automatically face punishing premiums—they can earn discounts through demonstrably safe driving. This creates a virtuous cycle where better driving leads to lower costs, fewer crashes, and ultimately, lives saved.
Perhaps most fascinating is CMT's breakthrough in real-time crash detection—something once deemed impossible using only smartphone sensors. Unlike solutions that focus only on catastrophic accidents, CMT's technology works across all crash severities, providing immediate assistance when needed most.
Looking toward the future, Balakrishnan shares his vision of data-driven road infrastructure improvements and the gradual integration of autonomous vehicles. All while maintaining a refreshingly human-centered approach to privacy: "Don't do anything with the data that you wouldn't want done with your own."
If you're interested in how technology can create safer roads while giving people more control over their driving costs and safety, this episode offers an inspiring glimpse into that future—one that's already arriving on roads worldwide.
Discover how technology is reshaping our lives and livelihoods.
Listen on: Apple Podcasts Spotify
More at https://linktr.ee/EvanKirstel
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
Hey everyone, fascinating topic today about
making the world's roads anddrivers safer with Cambridge
Mobile Telematics.
Hari, how are you I'm doinggreat?
Evan, how are you I'm doingwell?
Wonderful mission.
Thanks for joining to talkabout it.
Before that, maybe talk aboutyour bio and background the
journey from MIT research toreally global impact at CMT.
Speaker 2 (00:26):
Yeah, happy to do that.
I'm a computer scientist byprofession, professor of
computer science and AI at MIT.
I've been here a little over 25years.
About 20 years ago, with mycolleague, sam Madden, we
started a research projectcalled Cartel.
This was back before iPhonesand Androids.
I had this idea that we coulduse sensors that were becoming
(00:47):
more popular on mobile devicesto measure driving quality and
understand transportation,understand why our road
infrastructure has problems,understand why traffic problems
arose and, over time, also whycrashes were happening, and that
research project was hugelysuccessful in the academic
(01:07):
context.
We won a number of awards andthe press started writing about
some of the work we were doing.
One of them projects was thePothole Patrol, where, in 2007,
2008, we instrumented ourdevices with accelerometers and
GPS to produce every week aranking of the worst potholes in
(01:29):
the Boston area.
This got written up in theBoston Globe and the Wall Street
Journal, and I had this ideathat maybe we could use sensing
for social good, for solvingsocietal problems in
transportation.
I then met my co-founder, billPowers, who was CMT's CEO in
late 2009.
We started the company shortlythereafter, in 2010.
(01:50):
, and today we are the world'slargest telematics provider,
building mobile sensing andartificial intelligence
technologies to measure howpeople drive, to provide
incentives for better drivingworking by our partners in
insurance and ride shares andthe commercial space and then to
(02:10):
also detect crashes in realtime, when they happen,
regardless of severity andregardless of speed, and provide
assistance in the form ofroadside assistance or emergency
or towing and that's reallybeen the journey or emergency or
towing, and that's really beenthe journey.
We're currently about 475employees headquartered in
Cambridge, massachusetts, butreally a global company with
(02:34):
offices in many countries, andwe serve our users millions and
millions of users in about 25countries around the world.
Speaker 1 (02:39):
Amazing, and that journey sounds so
straightforward and simple andseamless.
Speaker 2 (02:44):
It was, it was, it was.
Speaker 1 (02:46):
From the early days doing mobile sensing experiments
it sounds like to truly globalplatform, I guess.
Billions of trips across 20 or30 countries that's
extraordinary.
Speaker 2 (02:58):
Thank you so much.
It was an overnight successthat took over 10 years.
So how did we scale?
I think, first and foremost, Idon't think there's a single
magic bullet secret, but firstand foremost, I would say that
we've always thought ofourselves as trying to
understand deeply the problemsthat our users and our customers
(03:19):
face and to try to solve thoseproblems with whatever
technology and products we couldbring.
So we've avoided the mindset ofpushing our products or our
technology onto people.
We really are trying to fall inlove with the problem and then
use our know-how and ouringenuity and our hard work
(03:42):
working again really in concertwith our partners to solve their
problems.
So I would say that we try tohelp our customers grow and
therefore grow ourselves.
That's probably been a keyaspect of what we've done Very
collaborative, it's.
One of our key values is to behighly collaborative, both
internally and with our partners, and try to focus on simple
(04:04):
solutions, try to come up withthe simplest way to tackle
problems, because these problemsare extremely complicated.
Speaker 1 (04:12):
Complicated indeed.
And you not only fuse IoT datafrom phones, but there's tags,
there's vehicles, dash cams, allother kinds of feeds.
How do you orchestrate all ofthese you know diverse data
streams into a single picture ofsafety?
Speaker 2 (04:30):
That's a great question.
It has been an evolution of howwe do that.
When we started, we were asingle data source.
We started with sensors onphones.
You know, every iPhone, androidphone out there for the last 15
years has come withaccelerometers and GPS and
gyroscopes and so on, and that'show we started.
Soon, we invented a small 5centimeter by 5 centimeter
(04:55):
device called the Tag.
That's an option.
It works together with a phone.
We've shipped over 40 millionTags around the world.
We've shipped over 40 milliontags around the world.
It's one of the most widely usedaftermarket automotive products
out there, purely for thepurpose of understanding,
driving and providing crashassistance and fulfilling our
mission of safety.
And then you know, we processdata from connected vehicles,
(05:19):
from vehicular dash cams and soon, and we've created this
platform.
We we call it the DriveWellFusion platform.
Drivewell is our suite ofproducts and Fusion is the
platform on which it runs.
Drivewell Fusion bringstogether data from a disparate,
wide variety of sources andharmonizes it, and the
(05:39):
harmonization means that,regardless of the type of data
source, we provide actionableinsights and output data that
allows for business applications, such as for pricing your
insurance or understanding yoursafety score or crash and claims
or behavior change to provideincentives for better driving.
Those business applicationsdon't have to depend on the
(06:01):
vagaries and the peculiaritiesand the differences of the data.
So it's not just that we makethe data schema look the same,
the data meaning is also thesame, and to do that requires a
lot of advances, both inunderstanding the physics of
movement patterns, the dynamicsof vehicular movement, as well
as the dynamics of how people,for example, use their phones
(06:23):
within the vehicle while they'redistracted.
There's patterns of movementinferred from the accelerometer
of the gyroscope of the phone,but it's really a physics part.
There's a signal processingpart and, of course, there's a
more modern machine learning andartificial intelligence
component.
So if we bring together physics, signal processing and machine
learning, we can do a lot betterthan with any one of those
(06:46):
methodologies alone.
And that's really been thesecret of our know-how as a team
that is equally adept atsolving difficult
physics-oriented problems orsignal processing problems as
with machine learning Amazing.
Speaker 1 (07:02):
And, unlike our friends at Big Tech, you really
advocate for a privacy-firstkind of consumer choice-led
model.
What does that mean in terms ofbalancing all that good, great
data you have with user trustand getting all those insights
on user behavior?
Speaker 2 (07:21):
Yeah, I won't speak for Big Tech, but what I will
say for us is there's a verysimple value principle Don't do
to or with the data what youdon't want to do with your own
data.
Imagine yourself you are a user.
If you're an engineer, aproduct person, an employee of
the company, don't do anythingwith the data that you don't
want to have happen to your owndata.
(07:42):
Everything stems from that.
We have a road safety board.
We've had a very strongindustry privacy board in the
past, led by some of the leadingthinkers at the intersection of
technology, privacy and law,and we've crafted a set of rules
that are simple to understand,and they all boil down to
exactly what I said Don't doanything that you don't want to
(08:02):
have happen to your own data.
Speaker 1 (08:06):
Well said, and you work with some very well-known,
successful insurance companies.
I won't name them all.
I certainly probably use one orone of them.
But how do those partnershipswork?
What do they look like and whathave they taught you about
behavior-driven technology?
Speaker 2 (08:25):
Yeah, we are very fortunate to work with some of
the largest and most successfuliconic, long-running companies.
These companies, you know, intech we have companies that are,
you know, 10 years old, 20years old, maybe 50 years old,
but in insurance we havecompanies that are 100 years old
and it is just.
I can't even imagine what itmust take to have a company that
lasts that long and thrives.
(08:46):
So we've been very fortunate towork with them and learn from
them.
The model is actually not thatcomplicated.
It's based on sharing a visionfor how we want to help them
succeed, and that is to come upwith a more predictive and more
equitable form of pricinginsurance.
(09:08):
And with telematics, for thefirst time in the history of
insurance, you have a way to setprices based on how people
drive and therefore provide morecontrol to our users, because
if you know how your insuranceprice is being set based on your
driving behavior, you canactually do something about it.
You can stop the phone whiledriving, you can stop
(09:28):
inattentive driving, you canreduce the amount of excessive
speeding, you can be moreattentive to.
For example, take a youngdriver 23 years old.
(09:56):
My daughters are in their early20s.
Traditionally, they would havehad a terrible insurance price
because, statistically, youngdrivers have a higher risk, but
not all young drivers have ahigher risk and, moreover, when
you give them the feedback andyou give them the incentives,
they can become better drivers.
And right now in my family, mytwo daughters are the best
drivers and that's becausethey're part of a behavioral
(10:17):
program that gives them rewardsfor better driving.
So all of a sudden, that highrisk has gone down.
It's good for them, it's goodfor a society with fewer crashes
, it's good for the insurancecompany because they now have,
you know, they can collect somepremium, they give them a break
on insurance but at the sametime they're going to hopefully
have fewer claims.
And certainly it's certainlygood for CMT as well, because we
(10:39):
achieve our mission, we driveup our growth and therefore it's
a win-win.
And the work is verypartnership oriented.
We understand the businessgoals of our insurance partners
or ride-share partners, forexample Uber, and we try to
solve those problems with themfor the end user.
So it's really trying to putthe end consumer first and
(11:03):
providing the right type ofincentive structure to make them
want to use it and then benefitfrom it.
Speaker 1 (11:09):
Wow, talk about a win-win-win type of incentive
structure to make them want touse it and then benefit from it.
Wow, talk about a win-win-win.
Let's talk a little bit moreabout the technology involved.
When was the sort of aha momentwhen you realized mobile
devices could become sort ofprecision safety tools?
Speaker 2 (11:24):
Yeah, that's a long time ago.
I wrote a paper with mycolleague Sam Madden and our
students back in.
We were working on this projectin 2004, 2005, 2006, and the
paper was published in 2006.
It's one of the more citedpapers in computer science.
It's called Cartel, a MobileSensing System, and there's a
(11:44):
sentence in the introductionthat says that soon these will
be available on smartphones andmobile devices.
And again just going back tothat time, it was predates, the
iPhone and the Android ecosystem.
It was Nokia phones which didhave sensors and I remember the
Motorola Razr phone that flipped.
I hear it's coming back now,but those were the phones that
(12:08):
we were looking at.
But to me I don't want to say itseemed obvious, but it seemed
to me inevitable that right fromthe late 90s, when again
predating Wi-Fi, it seemedinevitable to me that, given
technology trends, theminiaturization of computing was
happening at big scale.
(12:28):
And I got invited to someworkshops very early as in my
graduate school career, at theend of my graduate school career
, very early in my own career,where they were talking about
how sensors, mems technologies,were getting smaller and smaller
and more economical andwireless communication was
becoming more widespread andsmaller and smaller.
So when everything becomessmaller and smaller and faster,
(12:50):
it seemed inevitable to me thatthis would happen, that we would
be able to build devices andthe only question was whether we
could make them batteryefficient.
Because they were small, they'dbe powered by batteries and I
wouldn't say the only questionthere was the question of
battery efficiency, there was aquestion of cost and there was a
question of accuracy whetherthese consumer devices they're
(13:11):
super low cost.
And you have to remember, back25 years ago, sensors were used
in a lot of applications likeavionics, the military and
industrial process, and theywere really good sensors, but
they were expensive sensors.
People would pay $5,000,$10,000, not $5,000 or $10,000,
but $5,000 or $10,000 for theseexpensive sensors.
And so the question was whetherwe could get them to be
(13:35):
accurate enough at low cost andenergy efficient.
And our research and developmentwork at both at MIT and then at
CMT, as well as the entireindustry, really worked very
hard.
And, of course, the watershedmoment came when really the
iPhones and then Android.
Around that time that ecosystemexploded and when we started it
(13:58):
was not possible to havebackground sensing applications
run on iPhones.
So we built our first prototypeand research.
We did research on an oldjailbroken iPhone where we got
root access.
We bought an iPhone.
We just wanted to show that itcould work feasibility-wise and
we could, which meant that ifApple wanted, they could support
it, and we placed a bet that,yeah, you know what they
(14:21):
probably will, because there areso many useful things you could
do with it outside of roadsafety.
You could do health andwellness.
You could do things likenotifying people of you know
things that they need to do, andso on and so forth.
And then we did it with Androidin the first in 2012, when we
launched our first mobileprobably the world's first
mobile usage based insurancepilot program with a leading
(14:44):
insurer.
It was only available onAndroid.
It was only in 2013 that AppleiOS opened up this capability to
developers, but we were rightthere when that happened.
And then when we developed thetag, again based on a lot of
feedback from one of the earlystrategic partner discovery
(15:05):
insurer in South Africa, it wasreally a product that almost
looked impossible but ended upbeing possible.
And this was the firstcompletely battery-operated
small device that you could juststick to the windshield and
would have four years of batterylife.
At the time, a little bitbetter, even better now and the
(15:26):
cost has come down.
And again, the idea of takingthis type of telematic sensing
technology that was previouslyavailable in industrial
applications, that waspreviously available only to
large fleets, which were payinga lot of money, and making it
oriented toward consumerproducts so consumers could use
(15:46):
it easily and at really low cost.
That really was one of the keyinnovations.
And, of course, you don't endup with the most accurate
sensing data set, and that'swhere the technology comes in.
That's where the AI and thephysics and the machine learning
and the signal processing comein, which is we're able to
handle all of the errors in theraw data to produce really good,
(16:08):
accurate outputs.
Speaker 1 (16:11):
Amazing.
You also solved something thatwas said to be difficult, if not
impossible, early onidentifying crashes in real time
from mobile sensors.
Of course, recently we've seenApple baking that into the
iPhone and the watch, but whatenabled that breakthrough on
that end?
Speaker 2 (16:28):
That's a difficult question.
It's actually a problem.
It still remains a verydifficult problem to achieve
extremely high precision, whichis to say no false positives,
and high recall, which iscapturing every crash, and
there's a trade-off.
So, for example, ourtechnologies achieve we are
oriented toward high recall.
(16:48):
Our customers and our userswant to catch every possible
crash as possible and we focuson that.
Apple has it in the iPhone, onthe newer iPhones and the newer
watches, and that's focused ongetting the very highest impact,
life-threatening crashes,whereas ours is across the
spectrum, including roadsideassistance and claims using
insurance claims.
(17:08):
So it's just picking differentpoints in the spectrum.
I'll be honest, I thought backin 2014 that this could be
really difficult, if notimpossible, to solve on a pure
mobile phone device.
So one of the motivations forour tag was to develop a
windshield mounted device thatcould do it, and actually what's
interesting is that becausethat device is windshield
(17:29):
mounted and we have millions andmillions of them every day in
the field it's a little easierto solve the problem on a
windshield mounted devicebecause it's not being moved
around on the body of the personand it's not in your pocket
where it could get damped.
The signal could get damped, orsomebody could put it in a
handbag and leave it in thebackseat and maybe it doesn't
feel the shock the same way.
(17:50):
But we have this almost unfairadvantage, which is we have a
windshield mounted tag on 60% ofour users, which means and they
collect both the phone data andthe tag data, which has allowed
us to really build models thatwork with any data source,
because lots of our users areproviding us two data sources
that work with any data source,because lots of our users are
providing us two data sources.
Second, we have deeppartnerships where we're getting
(18:13):
ground truth on what actuallyhappened and we work with our
partners to provide them evenbetter results that are tuned to
their demographic.
Countries are different.
The type of driving isdifferent in different countries
, so we're able to then adaptour models.
We have a base set of modelsand then we're able to then
adapt our models.
We have a base set of modelsand then we're able to adapt
them to particular kind ofconditions, and all of that has
(18:34):
allowed us to succeed.
But I'll say this we believe weare best in class.
We have the most experience andthe most accurate outcomes, but
we're still working really hardon further improvements and
there have been breakthroughs inAI.
You know we now have modelsthat we use that are quite
interesting in that they havesome inspiration from take some
(18:57):
inspiration from the type ofmodels used in chat, gpt for
natural language.
So I think advances in AIcontinue to appear and we're
able to benefit from some ofthose advances in the research
literature as well.
Speaker 1 (19:10):
Fantastic and the impact is clear.
I mean, I see you're creditedwith helping prevent over 40,000
road injuries and fatalities.
That must be very gratifying,but at the same time, big
picture, at least here in the US, we have a pretty poor driving
safety record compared to ourpeer rich nations, let's say
really concerning fatality rates.
You know, rich nations, let'ssay really concerning fatality
(19:33):
rates.
What's your takeaway from that?
How can we or you impact evenmore change for good?
Speaker 2 (19:45):
Yeah, that's a great question.
I think our latest numbers arethat our analysis shows we've
averted over 90,000 crashes and50,000 injuries and fatalities,
and I think these are bignumbers, but they're not
anywhere near what we need to doat national or global scale,
and we're working hard with ourpartners on this, Okay, so what
do we need to do and why arethey problematic?
(20:06):
So I'll give you some bad newsand then some good news.
So the bad news, in some ways,is that since, especially since
2017, vehicles have come ladenwith more and more technologies.
A lot of it is around safetyand yet, if you look at the data
, for several years, crash rateshave not kept up in terms of,
they've not come down at thesame proportion.
(20:27):
I call that the vehicle paradox,and one of the reasons for that
is that the technology actuallyis quite complicated.
For example, you have a blindspot monitor, which is good.
You might have one of thosethings that keeps you in your
lane, but sometimes what happensis it pulls you.
You're trying to move lanes andit pulls you back and you're
wrestling with the car to stayin the lane.
So these technologies have morecomplicated human machine
(20:49):
interfaces, much more so thanthey used to be, and cars have
become quite different from eachother in terms of their
capability, and sometimes peoplewho drive two cars end up
getting confused.
So there are all these issuesthat I think are one of the
reasons why crash rates haven'tcome down as much.
And then we have phonedistraction, which is a huge
factor in terms of causingpeople to lose attention or not
(21:14):
pay attention and get intocrashes.
The US continues to have thehighest rate of phone
distraction per mile or per hourof driving compared to most
other countries.
It's a place where we're numberone when we shouldn't be, at
least among the countries we'vestudied.
But on the positive side, we'veseen about a 9% drop in the
amount of distracted driving perhour for users in our program
(21:37):
over the last year, and we thinkthat's because these programs
are now scaling.
There's lots of incentives inplace.
There's punishments in placewhere people's premiums might go
up if they get into troublebecause of crashes, and then
there are laws againstdistracted driving that have you
know that have come about inmany states.
I believe over 30 states nowhave laws against phone
(22:00):
distracted driving, and that andCMT's broader analytics studies
not, you know, looking atindividual users.
But just looking at overallrates suggest that at least in
the several months after lawspass they do help and of course
after that it depends on therate of enforcement.
So I think all of these things,there is some promise that the
(22:21):
distracted driving rate iscoming down.
I do think that these programs,when we see examples of
successes some of our partners,for example, Discover in South
Africa sees a 26% lower crashrate compared to the rest of the
population for people engagedwith the behavioral program.
In Japan, our customer IOE seesan 18% lower crash rate and
(22:43):
even in the US there aresuccessful examples amongst our
insurance partners where themajority of users in the program
see discounts.
In the program see discounts,the majority of users see
discounts.
The majority of users actuallyhave better behavior in terms of
their driving quality and lowercrash rates.
But you know we continue to dothe work because the overall
(23:04):
penetration in the markets forusers who use this on a daily
basis is still in the highsingle-digit percentages.
So we have a long way to go.
We've done well but we have along way to go.
Speaker 1 (23:16):
Long way to go, indeed.
And what does the next decadeof road safety look like?
So many emerging technologies,smart glasses,
vehicle-to-vehicle communication.
What else is on the radar thatwill make our road safer?
Speaker 2 (23:29):
Yeah, there's a lot, I think, going on.
I think, first and foremost, Ithink when we scale up the
telematics-based programs thatCMT works on, I think we will
see dramatic changes in humandriving behavior.
Despite the complexity ofvehicles, I think these programs
work.
I think we'll continue to see areduction in distracted driving
(23:50):
.
I really believe that.
I think there's a lot of dataavailable now, a lot of
incentives to prevent it andpenalties as well, so I think
that'll make a big difference.
Number two I mean I think theroad to autonomy will continue.
Robot taxis are coming soon.
I think that their penetrationinto the consumer driving market
(24:12):
will take more time and it'llbe a long, long haul because the
average age of a vehicle in theUS on the road is about over 12
years, so people are unlikelyto just replace their car to get
an autonomous vehicle.
But it'll happen over the next15 or so years, maybe 15 to 20
years, and then I think thatwill really be dramatic and in
(24:33):
that world I don't think crashesare going to be eliminated.
They're going to be lower butdifferent, and anyone doing
insurance or safety in thatworld will be fundamentally
based on telematics, because nolonger does it matter you know
the color of the skin, or Idon't think they use that in
pricing.
But the type of car you drive,the color of your car or the age
or demographic, the place youlive, may matter a little bit
(24:58):
because of crash rates, butwhether somebody is 20 years old
or 50 years old won't matter somuch.
But the way in which you ratedrivers will be fundamentally
based on telematics data the wayin which you rate the robotic
drivers, and that has to do withevaluating the quality of
software, the quality in whichyou rate the robotic drivers,
and that has to do withevaluating the quality of
software, the quality of thesensors and the quality of the
(25:18):
AI.
So we think CMT is wellpositioned 15 years from now.
In the interim time, one of theareas where we'll see dramatic
improvements is roadinfrastructure.
Data-driven decision-making inroad infrastructure is an area
that we're pioneering andworking on.
It's part of a product we havecalled Street Vision, which uses
(25:42):
aggregate, anonymized data toinform cities, municipalities
and states about how to improveroad infrastructure Placement of
stop signs, placement of speedbumps, bus stops, highway speed
limits, suburban roads, urbanroads.
How can you changeinfrastructure, both in the
short term to lower crashes, todetermine the effect of an
(26:04):
intervention before it happens.
It's very difficult to put astop sign, go about measuring it
and then deciding if it was agood idea or not.
What if we use data and AI tomake the determination before?
That's what strict vision is upto, and then, as we head on the
road to autonomy, we thinkinfrastructure has to change
that.
We think that there's going tobe infrastructural changes
(26:26):
needed to allow for thecoexistence of autonomous and
semi-autonomous and manualvehicles, and not to mention
just not just the vehiclesautonomous and semi-autonomous
and manual vehicles, and not tomention just not just the
vehicles.
We share roads with bicyclistsand motorcyclists and, you know,
pedestrians and runners andconstruction workers.
Everybody has to be safe.
This is important because, youknow, 50% of serious injuries on
(26:50):
roads and fatalities happen topeople outside the vehicle.
Serious injuries on roads andfatalities happen to people
outside the vehicle.
20% of such injuries andfatalities in the US are to
people outside the vehicle.
So they have to be part of therevolution, part of the
technological evolution, and Icould imagine a future where we
have clothing that we wear withsensors on them that keep us
(27:10):
safe on the road, and I thinkthat that's the future we're
headed together toward.
You know things like V2V, v2i.
I think these may be enablingtechnologies, but they're very
low-level communicationtechnologies.
I like to think top-down of theoutcomes we want to achieve,
and the outcomes we want toachieve are data-driven,
ai-driven and behavioral-driven.
Speaker 1 (27:34):
Wonderful.
Well, you're in the lovely cityof Cambridge, Massachusetts.
How's cycling in and aroundCambridge these days Getting
better, or is there yet work tobe done?
As you, and your team come intothe office.
Speaker 2 (27:45):
Much better than it used to be.
I ride my bike almost every dayto work, and Cambridge has done
a good job.
They've tried a number of goodexperiments.
For example, they have onewhere they put the bike lane
near the curb and they haveparking in the middle and then
they have cars on the other side.
That has two benefits Cyclistsdon't get doored as often and
(28:06):
second, you have a shieldbetween the bike lane and the
traffic.
They've tried a number ofexperiments, but, as with
anything, anything it's a workin progress because, uh, you
know, like with any societalproblem, there's people who like
bicyclists, people who don't,there are people who like
drivers, cars, they don't.
But ultimately we all have tocoexist.
There's only a certain numberof kilometers or miles of roads
(28:27):
and we have to share it.
So, um, you know, boston'sknown for some uh, let's just
say aggressive driving.
Speaker 1 (28:35):
Well, we're number one Of the worst drivers, that
is.
Speaker 2 (28:39):
I don't believe we're the worst drivers.
We have data showing where theworst drivers are Okay.
Speaker 1 (28:45):
We'll keep that data top secret, but thanks so much
for joining us Reallyeye-opening discussion.
Appreciate the mission andspending time here.
Thank you, evan, thanks so muchand thanks everyone for
listening, watching, sharingthis episode and check out our
new TV show, TechImpact TV, nowon Fox, Business and Bloomberg.
Thanks so much.
Thanks, Ari, Thank you.
Hey everyone, fascinating topic today about
making the world's roads anddrivers safer with Cambridge
Mobile Telematics.
Hari, how are you I'm doinggreat?
Evan, how are you I'm doingwell?
Wonderful mission.
Thanks for joining to talkabout it.
Before that, maybe talk aboutyour bio and background the
journey from MIT research toreally global impact at CMT.
Speaker 2 (00:26):
Yeah, happy to do that.
I'm a computer scientist byprofession, professor of
computer science and AI at MIT.
I've been here a little over 25years.
About 20 years ago, with mycolleague, sam Madden, we
started a research projectcalled Cartel.
This was back before iPhonesand Androids.
I had this idea that we coulduse sensors that were becoming
(00:47):
more popular on mobile devicesto measure driving quality and
understand transportation,understand why our road
infrastructure has problems,understand why traffic problems
arose and, over time, also whycrashes were happening, and that
research project was hugelysuccessful in the academic
(01:07):
context.
We won a number of awards andthe press started writing about
some of the work we were doing.
One of them projects was thePothole Patrol, where, in 2007,
2008, we instrumented ourdevices with accelerometers and
GPS to produce every week aranking of the worst potholes in
(01:29):
the Boston area.
This got written up in theBoston Globe and the Wall Street
Journal, and I had this ideathat maybe we could use sensing
for social good, for solvingsocietal problems in
transportation.
I then met my co-founder, billPowers, who was CMT's CEO in
late 2009.
We started the company shortlythereafter, in 2010.
(01:50):
, and today we are the world'slargest telematics provider,
building mobile sensing andartificial intelligence
technologies to measure howpeople drive, to provide
incentives for better drivingworking by our partners in
insurance and ride shares andthe commercial space and then to
(02:10):
also detect crashes in realtime, when they happen,
regardless of severity andregardless of speed, and provide
assistance in the form ofroadside assistance or emergency
or towing and that's reallybeen the journey or emergency or
towing, and that's really beenthe journey.
We're currently about 475employees headquartered in
Cambridge, massachusetts, butreally a global company with
(02:34):
offices in many countries, andwe serve our users millions and
millions of users in about 25countries around the world.
Speaker 1 (02:39):
Amazing, and that journey sounds so
straightforward and simple andseamless.
Speaker 2 (02:44):
It was, it was, it was.
Speaker 1 (02:46):
From the early days doing mobile sensing experiments
it sounds like to truly globalplatform, I guess.
Billions of trips across 20 or30 countries that's
extraordinary.
Speaker 2 (02:58):
Thank you so much.
It was an overnight successthat took over 10 years.
So how did we scale?
I think, first and foremost, Idon't think there's a single
magic bullet secret, but firstand foremost, I would say that
we've always thought ofourselves as trying to
understand deeply the problemsthat our users and our customers
(03:19):
face and to try to solve thoseproblems with whatever
technology and products we couldbring.
So we've avoided the mindset ofpushing our products or our
technology onto people.
We really are trying to fall inlove with the problem and then
use our know-how and ouringenuity and our hard work
(03:42):
working again really in concertwith our partners to solve their
problems.
So I would say that we try tohelp our customers grow and
therefore grow ourselves.
That's probably been a keyaspect of what we've done Very
collaborative, it's.
One of our key values is to behighly collaborative, both
internally and with our partners, and try to focus on simple
(04:04):
solutions, try to come up withthe simplest way to tackle
problems, because these problemsare extremely complicated.
Speaker 1 (04:12):
Complicated indeed.
And you not only fuse IoT datafrom phones, but there's tags,
there's vehicles, dash cams, allother kinds of feeds.
How do you orchestrate all ofthese you know diverse data
streams into a single picture ofsafety?
Speaker 2 (04:30):
That's a great question.
It has been an evolution of howwe do that.
When we started, we were asingle data source.
We started with sensors onphones.
You know, every iPhone, androidphone out there for the last 15
years has come withaccelerometers and GPS and
gyroscopes and so on, and that'show we started.
Soon, we invented a small 5centimeter by 5 centimeter
(04:55):
device called the Tag.
That's an option.
It works together with a phone.
We've shipped over 40 millionTags around the world.
We've shipped over 40 milliontags around the world.
It's one of the most widely usedaftermarket automotive products
out there, purely for thepurpose of understanding,
driving and providing crashassistance and fulfilling our
mission of safety.
And then you know, we processdata from connected vehicles,
(05:19):
from vehicular dash cams and soon, and we've created this
platform.
We we call it the DriveWellFusion platform.
Drivewell is our suite ofproducts and Fusion is the
platform on which it runs.
Drivewell Fusion bringstogether data from a disparate,
wide variety of sources andharmonizes it, and the
(05:39):
harmonization means that,regardless of the type of data
source, we provide actionableinsights and output data that
allows for business applications, such as for pricing your
insurance or understanding yoursafety score or crash and claims
or behavior change to provideincentives for better driving.
Those business applicationsdon't have to depend on the
(06:01):
vagaries and the peculiaritiesand the differences of the data.
So it's not just that we makethe data schema look the same,
the data meaning is also thesame, and to do that requires a
lot of advances, both inunderstanding the physics of
movement patterns, the dynamicsof vehicular movement, as well
as the dynamics of how people,for example, use their phones
(06:23):
within the vehicle while they'redistracted.
There's patterns of movementinferred from the accelerometer
of the gyroscope of the phone,but it's really a physics part.
There's a signal processingpart and, of course, there's a
more modern machine learning andartificial intelligence
component.
So if we bring together physics, signal processing and machine
learning, we can do a lot betterthan with any one of those
(06:46):
methodologies alone.
And that's really been thesecret of our know-how as a team
that is equally adept atsolving difficult
physics-oriented problems orsignal processing problems as
with machine learning Amazing.
Speaker 1 (07:02):
And, unlike our friends at Big Tech, you really
advocate for a privacy-firstkind of consumer choice-led
model.
What does that mean in terms ofbalancing all that good, great
data you have with user trustand getting all those insights
on user behavior?
Speaker 2 (07:21):
Yeah, I won't speak for Big Tech, but what I will
say for us is there's a verysimple value principle Don't do
to or with the data what youdon't want to do with your own
data.
Imagine yourself you are a user.
If you're an engineer, aproduct person, an employee of
the company, don't do anythingwith the data that you don't
want to have happen to your owndata.
(07:42):
Everything stems from that.
We have a road safety board.
We've had a very strongindustry privacy board in the
past, led by some of the leadingthinkers at the intersection of
technology, privacy and law,and we've crafted a set of rules
that are simple to understand,and they all boil down to
exactly what I said Don't doanything that you don't want to
(08:02):
have happen to your own data.
Speaker 1 (08:06):
Well said, and you work with some very well-known,
successful insurance companies.
I won't name them all.
I certainly probably use one orone of them.
But how do those partnershipswork?
What do they look like and whathave they taught you about
behavior-driven technology?
Speaker 2 (08:25):
Yeah, we are very fortunate to work with some of
the largest and most successfuliconic, long-running companies.
These companies, you know, intech we have companies that are,
you know, 10 years old, 20years old, maybe 50 years old,
but in insurance we havecompanies that are 100 years old
and it is just.
I can't even imagine what itmust take to have a company that
lasts that long and thrives.
(08:46):
So we've been very fortunate towork with them and learn from
them.
The model is actually not thatcomplicated.
It's based on sharing a visionfor how we want to help them
succeed, and that is to come upwith a more predictive and more
equitable form of pricinginsurance.
(09:08):
And with telematics, for thefirst time in the history of
insurance, you have a way to setprices based on how people
drive and therefore provide morecontrol to our users, because
if you know how your insuranceprice is being set based on your
driving behavior, you canactually do something about it.
You can stop the phone whiledriving, you can stop
(09:28):
inattentive driving, you canreduce the amount of excessive
speeding, you can be moreattentive to.
For example, take a youngdriver 23 years old.
(09:56):
My daughters are in their early20s.
Traditionally, they would havehad a terrible insurance price
because, statistically, youngdrivers have a higher risk, but
not all young drivers have ahigher risk and, moreover, when
you give them the feedback andyou give them the incentives,
they can become better drivers.
And right now in my family, mytwo daughters are the best
drivers and that's becausethey're part of a behavioral
(10:17):
program that gives them rewardsfor better driving.
So all of a sudden, that highrisk has gone down.
It's good for them, it's goodfor a society with fewer crashes
, it's good for the insurancecompany because they now have,
you know, they can collect somepremium, they give them a break
on insurance but at the sametime they're going to hopefully
have fewer claims.
And certainly it's certainlygood for CMT as well, because we
(10:39):
achieve our mission, we driveup our growth and therefore it's
a win-win.
And the work is verypartnership oriented.
We understand the businessgoals of our insurance partners
or ride-share partners, forexample Uber, and we try to
solve those problems with themfor the end user.
So it's really trying to putthe end consumer first and
(11:03):
providing the right type ofincentive structure to make them
want to use it and then benefitfrom it.
Speaker 1 (11:09):
Wow, talk about a win-win-win type of incentive
structure to make them want touse it and then benefit from it.
Wow, talk about a win-win-win.
Let's talk a little bit moreabout the technology involved.
When was the sort of aha momentwhen you realized mobile
devices could become sort ofprecision safety tools?
Speaker 2 (11:24):
Yeah, that's a long time ago.
I wrote a paper with mycolleague Sam Madden and our
students back in.
We were working on this projectin 2004, 2005, 2006, and the
paper was published in 2006.
It's one of the more citedpapers in computer science.
It's called Cartel, a MobileSensing System, and there's a
(11:44):
sentence in the introductionthat says that soon these will
be available on smartphones andmobile devices.
And again just going back tothat time, it was predates, the
iPhone and the Android ecosystem.
It was Nokia phones which didhave sensors and I remember the
Motorola Razr phone that flipped.
I hear it's coming back now,but those were the phones that
(12:08):
we were looking at.
But to me I don't want to say itseemed obvious, but it seemed
to me inevitable that right fromthe late 90s, when again
predating Wi-Fi, it seemedinevitable to me that, given
technology trends, theminiaturization of computing was
happening at big scale.
(12:28):
And I got invited to someworkshops very early as in my
graduate school career, at theend of my graduate school career
, very early in my own career,where they were talking about
how sensors, mems technologies,were getting smaller and smaller
and more economical andwireless communication was
becoming more widespread andsmaller and smaller.
So when everything becomessmaller and smaller and faster,
(12:50):
it seemed inevitable to me thatthis would happen, that we would
be able to build devices andthe only question was whether we
could make them batteryefficient.
Because they were small, they'dbe powered by batteries and I
wouldn't say the only questionthere was the question of
battery efficiency, there was aquestion of cost and there was a
question of accuracy whetherthese consumer devices they're
(13:11):
super low cost.
And you have to remember, back25 years ago, sensors were used
in a lot of applications likeavionics, the military and
industrial process, and theywere really good sensors, but
they were expensive sensors.
People would pay $5,000,$10,000, not $5,000 or $10,000,
but $5,000 or $10,000 for theseexpensive sensors.
And so the question was whetherwe could get them to be
(13:35):
accurate enough at low cost andenergy efficient.
And our research and developmentwork at both at MIT and then at
CMT, as well as the entireindustry, really worked very
hard.
And, of course, the watershedmoment came when really the
iPhones and then Android.
Around that time that ecosystemexploded and when we started it
(13:58):
was not possible to havebackground sensing applications
run on iPhones.
So we built our first prototypeand research.
We did research on an oldjailbroken iPhone where we got
root access.
We bought an iPhone.
We just wanted to show that itcould work feasibility-wise and
we could, which meant that ifApple wanted, they could support
it, and we placed a bet that,yeah, you know what they
(14:21):
probably will, because there areso many useful things you could
do with it outside of roadsafety.
You could do health andwellness.
You could do things likenotifying people of you know
things that they need to do, andso on and so forth.
And then we did it with Androidin the first in 2012, when we
launched our first mobileprobably the world's first
mobile usage based insurancepilot program with a leading
(14:44):
insurer.
It was only available onAndroid.
It was only in 2013 that AppleiOS opened up this capability to
developers, but we were rightthere when that happened.
And then when we developed thetag, again based on a lot of
feedback from one of the earlystrategic partner discovery
(15:05):
insurer in South Africa, it wasreally a product that almost
looked impossible but ended upbeing possible.
And this was the firstcompletely battery-operated
small device that you could juststick to the windshield and
would have four years of batterylife.
At the time, a little bitbetter, even better now and the
(15:26):
cost has come down.
And again, the idea of takingthis type of telematic sensing
technology that was previouslyavailable in industrial
applications, that waspreviously available only to
large fleets, which were payinga lot of money, and making it
oriented toward consumerproducts so consumers could use
(15:46):
it easily and at really low cost.
That really was one of the keyinnovations.
And, of course, you don't endup with the most accurate
sensing data set, and that'swhere the technology comes in.
That's where the AI and thephysics and the machine learning
and the signal processing comein, which is we're able to
handle all of the errors in theraw data to produce really good,
(16:08):
accurate outputs.
Speaker 1 (16:11):
Amazing.
You also solved something thatwas said to be difficult, if not
impossible, early onidentifying crashes in real time
from mobile sensors.
Of course, recently we've seenApple baking that into the
iPhone and the watch, but whatenabled that breakthrough on
that end?
Speaker 2 (16:28):
That's a difficult question.
It's actually a problem.
It still remains a verydifficult problem to achieve
extremely high precision, whichis to say no false positives,
and high recall, which iscapturing every crash, and
there's a trade-off.
So, for example, ourtechnologies achieve we are
oriented toward high recall.
(16:48):
Our customers and our userswant to catch every possible
crash as possible and we focuson that.
Apple has it in the iPhone, onthe newer iPhones and the newer
watches, and that's focused ongetting the very highest impact,
life-threatening crashes,whereas ours is across the
spectrum, including roadsideassistance and claims using
insurance claims.
(17:08):
So it's just picking differentpoints in the spectrum.
I'll be honest, I thought backin 2014 that this could be
really difficult, if notimpossible, to solve on a pure
mobile phone device.
So one of the motivations forour tag was to develop a
windshield mounted device thatcould do it, and actually what's
interesting is that becausethat device is windshield
(17:29):
mounted and we have millions andmillions of them every day in
the field it's a little easierto solve the problem on a
windshield mounted devicebecause it's not being moved
around on the body of the personand it's not in your pocket
where it could get damped.
The signal could get damped, orsomebody could put it in a
handbag and leave it in thebackseat and maybe it doesn't
feel the shock the same way.
(17:50):
But we have this almost unfairadvantage, which is we have a
windshield mounted tag on 60% ofour users, which means and they
collect both the phone data andthe tag data, which has allowed
us to really build models thatwork with any data source,
because lots of our users areproviding us two data sources
that work with any data source,because lots of our users are
providing us two data sources.
Second, we have deeppartnerships where we're getting
(18:13):
ground truth on what actuallyhappened and we work with our
partners to provide them evenbetter results that are tuned to
their demographic.
Countries are different.
The type of driving isdifferent in different countries
, so we're able to then adaptour models.
We have a base set of modelsand then we're able to then
adapt our models.
We have a base set of modelsand then we're able to adapt
them to particular kind ofconditions, and all of that has
(18:34):
allowed us to succeed.
But I'll say this we believe weare best in class.
We have the most experience andthe most accurate outcomes, but
we're still working really hardon further improvements and
there have been breakthroughs inAI.
You know we now have modelsthat we use that are quite
interesting in that they havesome inspiration from take some
(18:57):
inspiration from the type ofmodels used in chat, gpt for
natural language.
So I think advances in AIcontinue to appear and we're
able to benefit from some ofthose advances in the research
literature as well.
Speaker 1 (19:10):
Fantastic and the impact is clear.
I mean, I see you're creditedwith helping prevent over 40,000
road injuries and fatalities.
That must be very gratifying,but at the same time, big
picture, at least here in the US, we have a pretty poor driving
safety record compared to ourpeer rich nations, let's say
really concerning fatality rates.
You know, rich nations, let'ssay really concerning fatality
(19:33):
rates.
What's your takeaway from that?
How can we or you impact evenmore change for good?
Speaker 2 (19:45):
Yeah, that's a great question.
I think our latest numbers arethat our analysis shows we've
averted over 90,000 crashes and50,000 injuries and fatalities,
and I think these are bignumbers, but they're not
anywhere near what we need to doat national or global scale,
and we're working hard with ourpartners on this, Okay, so what
do we need to do and why arethey problematic?
(20:06):
So I'll give you some bad newsand then some good news.
So the bad news, in some ways,is that since, especially since
2017, vehicles have come ladenwith more and more technologies.
A lot of it is around safetyand yet, if you look at the data
, for several years, crash rateshave not kept up in terms of,
they've not come down at thesame proportion.
(20:27):
I call that the vehicle paradox,and one of the reasons for that
is that the technology actuallyis quite complicated.
For example, you have a blindspot monitor, which is good.
You might have one of thosethings that keeps you in your
lane, but sometimes what happensis it pulls you.
You're trying to move lanes andit pulls you back and you're
wrestling with the car to stayin the lane.
So these technologies have morecomplicated human machine
(20:49):
interfaces, much more so thanthey used to be, and cars have
become quite different from eachother in terms of their
capability, and sometimes peoplewho drive two cars end up
getting confused.
So there are all these issuesthat I think are one of the
reasons why crash rates haven'tcome down as much.
And then we have phonedistraction, which is a huge
factor in terms of causingpeople to lose attention or not
(21:14):
pay attention and get intocrashes.
The US continues to have thehighest rate of phone
distraction per mile or per hourof driving compared to most
other countries.
It's a place where we're numberone when we shouldn't be, at
least among the countries we'vestudied.
But on the positive side, we'veseen about a 9% drop in the
amount of distracted driving perhour for users in our program
(21:37):
over the last year, and we thinkthat's because these programs
are now scaling.
There's lots of incentives inplace.
There's punishments in placewhere people's premiums might go
up if they get into troublebecause of crashes, and then
there are laws againstdistracted driving that have you
know that have come about inmany states.
I believe over 30 states nowhave laws against phone
(22:00):
distracted driving, and that andCMT's broader analytics studies
not, you know, looking atindividual users.
But just looking at overallrates suggest that at least in
the several months after lawspass they do help and of course
after that it depends on therate of enforcement.
So I think all of these things,there is some promise that the
(22:21):
distracted driving rate iscoming down.
I do think that these programs,when we see examples of
successes some of our partners,for example, Discover in South
Africa sees a 26% lower crashrate compared to the rest of the
population for people engagedwith the behavioral program.
In Japan, our customer IOE seesan 18% lower crash rate and
(22:43):
even in the US there aresuccessful examples amongst our
insurance partners where themajority of users in the program
see discounts.
In the program see discounts,the majority of users see
discounts.
The majority of users actuallyhave better behavior in terms of
their driving quality and lowercrash rates.
But you know we continue to dothe work because the overall
(23:04):
penetration in the markets forusers who use this on a daily
basis is still in the highsingle-digit percentages.
So we have a long way to go.
We've done well but we have along way to go.
Speaker 1 (23:16):
Long way to go, indeed.
And what does the next decadeof road safety look like?
So many emerging technologies,smart glasses,
vehicle-to-vehicle communication.
What else is on the radar thatwill make our road safer?
Speaker 2 (23:29):
Yeah, there's a lot, I think, going on.
I think, first and foremost, Ithink when we scale up the
telematics-based programs thatCMT works on, I think we will
see dramatic changes in humandriving behavior.
Despite the complexity ofvehicles, I think these programs
work.
I think we'll continue to see areduction in distracted driving
(23:50):
.
I really believe that.
I think there's a lot of dataavailable now, a lot of
incentives to prevent it andpenalties as well, so I think
that'll make a big difference.
Number two I mean I think theroad to autonomy will continue.
Robot taxis are coming soon.
I think that their penetrationinto the consumer driving market
(24:12):
will take more time and it'llbe a long, long haul because the
average age of a vehicle in theUS on the road is about over 12
years, so people are unlikelyto just replace their car to get
an autonomous vehicle.
But it'll happen over the next15 or so years, maybe 15 to 20
years, and then I think thatwill really be dramatic and in
(24:33):
that world I don't think crashesare going to be eliminated.
They're going to be lower butdifferent, and anyone doing
insurance or safety in thatworld will be fundamentally
based on telematics, because nolonger does it matter you know
the color of the skin, or Idon't think they use that in
pricing.
But the type of car you drive,the color of your car or the age
or demographic, the place youlive, may matter a little bit
(24:58):
because of crash rates, butwhether somebody is 20 years old
or 50 years old won't matter somuch.
But the way in which you ratedrivers will be fundamentally
based on telematics data the wayin which you rate the robotic
drivers, and that has to do withevaluating the quality of
software, the quality in whichyou rate the robotic drivers,
and that has to do withevaluating the quality of
software, the quality of thesensors and the quality of the
(25:18):
AI.
So we think CMT is wellpositioned 15 years from now.
In the interim time, one of theareas where we'll see dramatic
improvements is roadinfrastructure.
Data-driven decision-making inroad infrastructure is an area
that we're pioneering andworking on.
It's part of a product we havecalled Street Vision, which uses
(25:42):
aggregate, anonymized data toinform cities, municipalities
and states about how to improveroad infrastructure Placement of
stop signs, placement of speedbumps, bus stops, highway speed
limits, suburban roads, urbanroads.
How can you changeinfrastructure, both in the
short term to lower crashes, todetermine the effect of an
(26:04):
intervention before it happens.
It's very difficult to put astop sign, go about measuring it
and then deciding if it was agood idea or not.
What if we use data and AI tomake the determination before?
That's what strict vision is upto, and then, as we head on the
road to autonomy, we thinkinfrastructure has to change
that.
We think that there's going tobe infrastructural changes
(26:26):
needed to allow for thecoexistence of autonomous and
semi-autonomous and manualvehicles, and not to mention
just not just the vehiclesautonomous and semi-autonomous
and manual vehicles, and not tomention just not just the
vehicles.
We share roads with bicyclistsand motorcyclists and, you know,
pedestrians and runners andconstruction workers.
Everybody has to be safe.
This is important because, youknow, 50% of serious injuries on
(26:50):
roads and fatalities happen topeople outside the vehicle.
Serious injuries on roads andfatalities happen to people
outside the vehicle.
20% of such injuries andfatalities in the US are to
people outside the vehicle.
So they have to be part of therevolution, part of the
technological evolution, and Icould imagine a future where we
have clothing that we wear withsensors on them that keep us
(27:10):
safe on the road, and I thinkthat that's the future we're
headed together toward.
You know things like V2V, v2i.
I think these may be enablingtechnologies, but they're very
low-level communicationtechnologies.
I like to think top-down of theoutcomes we want to achieve,
and the outcomes we want toachieve are data-driven,
ai-driven and behavioral-driven.
Speaker 1 (27:34):
Wonderful.
Well, you're in the lovely cityof Cambridge, Massachusetts.
How's cycling in and aroundCambridge these days Getting
better, or is there yet work tobe done?
As you, and your team come intothe office.
Speaker 2 (27:45):
Much better than it used to be.
I ride my bike almost every dayto work, and Cambridge has done
a good job.
They've tried a number of goodexperiments.
For example, they have onewhere they put the bike lane
near the curb and they haveparking in the middle and then
they have cars on the other side.
That has two benefits Cyclistsdon't get doored as often and
(28:06):
second, you have a shieldbetween the bike lane and the
traffic.
They've tried a number ofexperiments, but, as with
anything, anything it's a workin progress because, uh, you
know, like with any societalproblem, there's people who like
bicyclists, people who don't,there are people who like
drivers, cars, they don't.
But ultimately we all have tocoexist.
There's only a certain numberof kilometers or miles of roads
(28:27):
and we have to share it.
So, um, you know, boston'sknown for some uh, let's just
say aggressive driving.
Speaker 1 (28:35):
Well, we're number one Of the worst drivers, that
is.
Speaker 2 (28:39):
I don't believe we're the worst drivers.
We have data showing where theworst drivers are Okay.
Speaker 1 (28:45):
We'll keep that data top secret, but thanks so much
for joining us Reallyeye-opening discussion.
Appreciate the mission andspending time here.
Thank you, evan, thanks so muchand thanks everyone for
listening, watching, sharingthis episode and check out our
new TV show, TechImpact TV, nowon Fox, Business and Bloomberg.
Thanks so much.
Thanks, Ari, Thank you.
Popular Podcasts
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.